Commit 0a482342 authored by Alessandro Rubini's avatar Alessandro Rubini

doc: timer_ms is zero by default, interrupts are there

Signed-off-by: Alessandro Rubini's avatarAlessandro Rubini <rubini@gnudd.com>
parent 50929814
......@@ -529,12 +529,11 @@ The following parameters are used:
@item timer_ms=
The period of the internal timer. The timer is used to poll
for input events. We currently have no interrupt support so
we must poll, but the parameter will remain even when interrupt
is available, to disable it on request and reuse the current
polling code. The interval by default is 10ms and currently
only one timestamp is retrieved at each timer execution.
The period of the internal timer, if not zero.
The timer is used to poll for input events instead of enabling
the interrupt. The default interval is 0, which means to
use interrupt support. You may want to use the timer while
porting to a different carrier, before sorting out IRQ issues.
@item calib_s=
......@@ -585,7 +584,6 @@ leaving the other ones alone:
@smallexample
spusa.root# insmod fmc-fine-delay.ko busid=0x0200
[ 1336.532785] fmc_fine_delay: using a timer for input stamps (10 ms)
[ 1336.539021] spec 0000:02:00.0: reprogramming with fmc/fine-delay.bin
[ 1336.771452] spec 0000:02:00.0: FPGA programming successful
[ 1336.777063] spec 0000:02:00.0: Gateware successfully loaded
......@@ -598,7 +596,8 @@ spusa.root# insmod fmc-fine-delay.ko busid=0x0200
[ 1339.893847] fd_calibrate_outputs: ch4: 8ns @@859 (f 827, off 32, t 71.00)
[ 1339.962600] fmc_fine_delay: Found i2c device at 0x50
[ 1339.989642] fd_i2c_init: calibration on eeprom is invalid
[ 1339.995394] spec 0000:04:00.0: not using "fmc_fine_delay" according to modparam
[ 1339.995785] fmc_fine_delay: using interrupts for input
[ 1340.055394] spec 0000:04:00.0: not using "fmc_fine_delay" according to modparam
@end smallexample
If you use @code{show_sdb=1}, you'll get the following dump of the
......@@ -1763,9 +1762,6 @@ allows:
@itemize @bullet
@item We should use interrupts. The input is currently performed with
a kernel timer.
@item Calibration information in the EEPROM is not fixed for
endianness, so it only works on hosts of the same endianness as the
one where it has been programmed.
......
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment